proptesti {rigr}

Test of proportions from summary statistics

Description

Performs a one- or two-sample test of proportions using counts of successes and trials, rather than binary data. This test can be approximate or exact.

Usage

proptesti(
  x1,
  n1,
  x2 = NULL,
  n2 = NULL,
  exact = FALSE,
  null.hypoth = ifelse(is.null(x2) && is.null(n2), 0.5, 0),
  conf.level = 0.95,
  alternative = "two.sided",
  correct = FALSE,
  more.digits = 0
)

Arguments

x1

Number of successes in first sample

n1

Number of trials in first sample

x2

Number of successes in second sample

n2

Number of trials in second sample

exact

If true, performs a test of equality of proportions with Exact Binomial based confidence intervals.

null.hypoth

a number specifying the null hypothesis for the mean (or difference in means if performing a two-sample test). Defaults to 0.5 for one-sample and 0 for two-sample.

conf.level

confidence level of the test. Defaults to 0.95

alternative

a string: one of "less", "two.sided", or "greater" specifying the form of the test. Defaults to a two-sided test. When either "less" or "greater" is used, the corresponding one-sided confidence interval is returned.

correct

a logical indicating whether to perform a continuity correction

more.digits

a numeric value specifying whether or not to display more or fewer digits in the output. Non-integers are automatically rounded down.

Details

If x2 or n2 are specified, then both must be specified, and a two-sample test is run.

Value

A list of class proptesti. The print method lays out the information in an easy-to-read format.

tab

A formatted table of descriptive and inferential results (total number of observations, sample proportion, standard error of the proportion estimate), along with a confidence interval for the underlying proportion.

zstat

the value of the test statistic, if using an approximate test.

pval

the p-value for the test

par

A vector of information about the type of test (null hypothesis, alternative hypothesis, etc.)

Examples

# Two-sample test
proptesti(10, 100, 15, 200, alternative = "less")